I've been practising triad pair exercises and I really like the sound. I was noodling and came up with the attached pattern. It consists of alternating minor and major triads, implying a sequence of iv6 - V - i chords, where the i at the end of one sequence becomes the iv6 at the beginning of the next.
Is there a better harmonic analysis of what's going on here?
You've stumbled on a (fairly good) circle of fifths progression, especially one that sweeps through all 12 minor keys in the circle.
Your chord analysis of each individual unit of the progression is fine enough - what I'm seeing is iv-V-i=iv/v.
Circle of fifths progressions that only sweep through some of the 12 major/minor keys in the circle - or even alternate between major and minor keys in a I-vi-IV=I/IV pattern - are often found in the development sections of sonata-allegros.
Your harmonic analysis makes sense.
Personally, I practice a lot of sequential patterns like this, and when I do that I usually permute them through all inversions/chord positions. With that inversion process in mind, if you work through all the basic inversions/chord positions, you can do it with smooth voice leading, and good tendency tone treatment, with the elided chord (imin6=ivmin6), except in one place...
The inversion figures like 6/4/3 apply only to the right hand chord position. The figures like ^4 are scale degrees relative to the local key of each sequential iteration.
So, starting with the first iteration, in Cm, the top voice is scale degree ^4, and when it moves to the second iteration, using the next chord position, it resolves down as expected to ^♭3 of Cm, which then becomes ^♭6 in Gm.
Regarding the inversions/chord positions of the left hand, the pattern is simple start with the root in the top voice, then with each sequential iteration change the top voice to the chord's third, fifth, and added sixth. When you do that over the given root progression, the top line follows good voice leading, and as a practice pattern we get all the basic chord positions. Except in the third iteration.
In the third iteration the top voice moves from local tonic to leading tone. If you followed the pattern strictly, the leading tone will not resolve up. If you want the leading tone to resolve and extra chord can be inserted.
Someone will probably write a comment that says you don't need to resolve the leading tone, or that I am say you must resolve a leading tone. I'm not. I'm only pointing out the voice leading and tendency tone possibilities.
Your original question was...
Is there a better harmonic analysis of what's going on here?
Limited to only harmonic analysis I don't see a better way to look at what you analyzed.
But I suggest additionally analyzing the voice leading and scale degree tendencies, because that will allow you to take the simple harmonic elements and expand them into melodic figuration patterns.
The reason I reduced your original pattern to block chords in whole notes was to make it easier to see the voice leading actions. After that becomes clear, it's easier to realize the harmonic pattern with endless broken chord and embellished figuration with many different rhythms, contours, and meters.
I suppose if the ^1 ^7 ^1 top line was desired, it makes more sense to start the whole thing with position 6/5/3 and ending with 7/5/3 for the ^1 ^7 ^1 top line. Then that whole larger pattern would be repeated three times sequentially to hit all twelve tonics...
iv>V>i(new iv) is indeed the sequence, muddied a little with enharmonic triads written. And with 12 bars, we're brought through the whole cycle, and end up where we started, on G.
Or - we could start at the second half of bar 1. I>iv>V, which then takes us on to (bar 3) i, becoming then the parallel of I - i. There are most likely other patterns to be gleaned from sequences such as this. And other sequences made similarly!
